multi-data analysis based proactive defect detection and resolution may include analyzing operational data for an application to determine whether a functionality related to the application is below a predetermined threshold associated with the functionality related to the application, and based on the analysis, generating an indication to perform defect analysis related to the functionality related to the application. A sentiment analysis may be performed on consumer data related to the application to determine a sentiment of the consumer data related to the application, and a natural language processing (NLP) analysis may be performed on the consumer data related to the application to determine a function associated with a negative sentiment. application code and process data related to the application may be analyzed to determine a defect associated with the application. Further, a code of the application may be modified to correct the defect associated with the application.
|
16. A method for multi-data analysis based proactive defect detection and resolution, the method comprising:
performing a sentiment analysis on consumer data related to an application to determine a sentiment of the consumer data related to the application;
performing a language processing analysis, in response to a determination that the sentiment is a negative sentiment, on the consumer data related to the application to determine a function associated with the negative sentiment;
analyzing, in response to the determination that the sentiment is the negative sentiment, application code and process data related to the application to determine a defect associated with the application by
comparing a new user interaction pattern with the application to a previous user interaction pattern with the application, and
in response to a determination that the new user interaction pattern with the application is different from the previous user interaction pattern with the application, identifying the defect associated with the application based on the difference of new user interaction pattern with the application from the previous user interaction pattern with the application; and
modifying a code of the application to correct the defect associated with the application.
20. A non-transitory computer readable medium having stored thereon machine readable instructions for multi-data analysis based proactive defect detection and resolution, the machine readable instructions when executed cause at least one hardware processor to:
perform, by a data analyzer that is executed by at least one hardware processor,
a sentiment analysis on consumer data related to an application to determine a sentiment of the consumer data related to the application, and
a language processing analysis on the consumer data related to the application to determine a function associated with the sentiment;
analyze, by a defect detector that is executed by the at least one hardware processor, application code and process data related to the application to determine a defect associated with the application by
comparing a new user interaction pattern with the application to a previous user interaction pattern with the application to identify the defect associated with the application;
determine, by a defect ticket generator that is executed by the at least one hardware processor, a context of the defect associated with the application; and
generate, by the defect ticket generator, a defect ticket that includes an indication of the context of the defect associated with the application.
1. A multi-data analysis based proactive defect detection and resolution system comprising:
a data analyzer, executed by at least one hardware processor, to
perform
a sentiment analysis on consumer data related to an application to determine a sentiment of the consumer data related to the application, and
a natural language processing (NLP) analysis, in response to a determination that the sentiment is a negative sentiment, on the consumer data related to the application to determine a function associated with the negative sentiment;
a defect detector, executed by the at least one hardware processor, to
analyze, in response to the determination that the sentiment is the negative sentiment, application code and process data related to the application to determine a defect associated with the application by
comparing a new user interaction pattern with the application to a previous user interaction pattern with the application, and
in response to a determination that the new user interaction pattern with the application is different from the previous user interaction pattern with the application, identifying the defect associated with the application; and
a defect resolver, executed by the at least one hardware processor, to
modify a code of the application to correct the defect associated with the application.
2. The multi-data analysis based proactive defect detection and resolution system according to
a defect ticket generator, executed by the at least one hardware processor, to
determine a context of the defect associated with the application,
determine an entity associated with development of the function associated with the negative sentiment, and
generate a defect ticket that includes an indication of the context of the defect associated with the application, and the entity associated with development of the function associated with the negative sentiment.
3. The multi-data analysis based proactive defect detection and resolution system according to
4. The multi-data analysis based proactive defect detection and resolution system according to
comparing a text of the consumer data related to the application to a catalog of known functions related to the application, and
based on a match of the text of the consumer data related to the application to a catalog function from the catalog of known functions related to the application, identifying the catalog function as the function associated with the negative sentiment.
5. The multi-data analysis based proactive defect detection and resolution system according to
comparing a text of the consumer data related to the application to a catalog of known functions related to the application, and
based on a match of the text of the consumer data related to the application to a plurality of catalog functions from the catalog of known functions related to the application, identifying a catalog function from the plurality of catalog functions that includes a highest number of matches as the function associated with the negative sentiment.
6. The multi-data analysis based proactive defect detection and resolution system according to
generating a plurality of user-level graphs related to interactions of a plurality of users with the application,
merging the plurality of user-level graphs into an application-level graph for the application,
comparing the application-level graph for the application to a previous application-level graph for the application to determine a change between the application-level graphs for the application, and
determining the defect associated with the application based on the determined change between the application-level graphs for the application.
7. The multi-data analysis based proactive defect detection and resolution system according to
implementing an alternate functionality related to the application.
8. The multi-data analysis based proactive defect detection and resolution system according to
modifying code associated with a functionality related to the application.
9. The multi-data analysis based proactive defect detection and resolution system according to
determine a criticality level of the defect based on an analysis of a number of users who have downloaded a new version of the application versus an overall number of users who have downloaded the application,
in response to a determination that the criticality level of the defect is less than a predetermined criticality level threshold, assign a first time period for modifying the code of the application to correct the defect associated with the application, and
in response to a determination that the criticality level of the defect is greater than the predetermined criticality level threshold, assign a second time period for modifying the code of the application to correct the defect associated with the application, wherein the first time period is greater than the second time period.
10. The multi-data analysis based proactive defect detection and resolution system according to
determine a criticality level of the defect based on an analysis of a number of users who have updated to a new version of the application versus an overall number of users who have downloaded the application,
in response to a determination that the criticality level of the defect is less than a predetermined criticality level threshold, assign a first time period for modifying the code of the application to correct the defect associated with the application, and
in response to a determination that the criticality level of the defect is greater than the predetermined criticality level threshold, assign a second time period for modifying the code of the application to correct the defect associated with the application, wherein the first time period is greater than the second time period.
11. The multi-data analysis based proactive defect detection and resolution system according to
12. The multi-data analysis based proactive defect detection and resolution system according to
13. The multi-data analysis based proactive defect detection and resolution system according to
14. The multi-data analysis based proactive defect detection and resolution system according to
15. The multi-data analysis based proactive defect detection and resolution system according to
the danger score of less than a predetermined first danger score threshold represents no potential issues with the particular release version of the application,
the danger score of greater than the predetermined first danger score threshold and less than a predetermined second danger score threshold represents possible potential issues with the particular release version of the application, and
the danger score of greater than the predetermined second danger score threshold represents definite issues with the particular release version of the application.
17. The method for multi-data analysis based proactive defect detection and resolution according to
determining a context of the defect associated with the application;
determining an entity associated with development of the function associated with the negative sentiment; and
generating a defect ticket that includes an indication of the context of the defect associated with the application, and the entity associated with development of the function associated with the negative sentiment.
18. The method for multi-data analysis based proactive defect detection and resolution according to
comparing a text of the consumer data related to the application to a catalog of known functions related to the application; and
based on a match of the text of the consumer data related to the application to a catalog function from the catalog of known functions related to the application, identifying the catalog function as the function associated with the negative sentiment.
19. The method for multi-data analysis based proactive defect detection and resolution according to
comparing a text of the consumer data related to the application to a catalog of known functions related to the application; and
based on a match of the text of the consumer data related to the application to a plurality of catalog functions from the catalog of known functions related to the application, identifying a catalog function from the plurality of catalog functions that includes a highest number of matches as the function associated with the negative sentiment.
|
The present application is a Continuation of commonly assigned and co-pending U.S. patent application Ser. No. 14/961,328, filed Dec. 7, 2015, which claims priority to Indian patent application number 4153/CHE/2015, having a filing date of Aug. 10, 2015, the disclosures of which are hereby incorporated by reference in their entireties.
Applications, such as digital applications, may include any type of computer application that is designed to perform a task. For example, an automobile model generation application may be designed to generate a model of an automobile based on a user configuration of the automobile. Such applications may include defects that impact the intended functionality thereof.
Features of the present disclosure are illustrated by way of examples shown in the following figures. In the following figures, like numerals indicate like elements, in which:
For simplicity and illustrative purposes, the present disclosure is described by referring mainly to examples thereof. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present disclosure. It will be readily apparent however, that the present disclosure may be practiced without limitation to these specific details. In other instances, some methods and structures have not been described in detail so as not to unnecessarily obscure the present disclosure.
Throughout the present disclosure, the terms “a” and “an” are intended to denote at least one of a particular element. As used herein, the term “includes” means includes but not limited to, the term “including” means including but not limited to. The term “based on” means based at least in part on.
During development of an application, such as a digital application, the application may be tested to ensure that defects are addressed prior to deployment thereof. Once an application is deployed, for example, in an enterprise environment, a known group of users (i.e., consumers) may utilize the application. For example, the known group of users may download the application on their mobile devices or other devices, or otherwise use the application via their mobile devices or other devices without downloading. If a defect is encountered in the application, these users may utilize a predetermined channel to create a defect ticket or bug report, and send the defect ticket or bug report to the entity that developed the application, or to an entity that is tasked to resolve defects related to the application. Examples of defects may include a user's inability to log-in, view a menu, etc., and generally, a user's inability to perform any type of functionality that is supposed to be performed by the application. Based on the relevant service level agreements (SLAs) related to the application, the defect is typically fixed, and the fixed application may be re-deployed.
For example, referring to
In the digital world outside of an enterprise environment that includes a known group of users, or even in certain enterprise environments that may include a known group of users, the model of
In order to address the aforementioned aspects related to detection and resolution of defects in the digital world outside of an enterprise environment that includes a known group of users, or even in certain enterprise environments that may include a known or an unknown group of users, a multi-data analysis based proactive defect detection and resolution system, and a method for multi-data analysis based proactive defect detection and resolution are disclosed. The system and method disclosed herein may comprise sentiment analysis, natural language processing (NLP) techniques, and analysis of event flow data to determine potential defects, and to trigger a defect resolution process. The sentiment analysis may represent, for example, a positive or negative sentiment of user feedback related to an application. The NLP techniques as described herein may be used to determine functionality of an application associated with a particular user feedback, and the basis for a particular sentiment determination. The event flow data may represent a sequence of functionalities of an application implemented or otherwise encountered by a user (e.g., login, menu, logout, etc., functionalities implemented or otherwise encountered).
The system and method disclosed herein may include the analysis of multiple types of data from various sources to detect defects, and to determine appropriate actions that are to be taken to resolve the defects. According to an example, the system and method disclosed herein may be deployed as a service to utilize data sources such as, for example, application store feedback, release history data, event flow data, and software (i.e., machine readable instructions) engineering data (i.e., application code development data as described herein), and by using application programming interfaces (APIs), the system and method disclosed herein may generate insights and actions to resolve the defects, as well as to provide further guidance for development of an application. The system and method disclosed herein may thus provide a link between data related to consumers that utilize an application and provide feedback, for example, at an application store, data related to operation of the application (e.g., process data), and data related to development of the application and fixing of defects related to the application (i.e., development data). The system and method disclosed herein may also provide for the derivation of insights into the health of an enterprise's systems and processes, and triggering of contextual actions through analytics of multi-dimensional data.
According to an example, system and method disclosed herein may determine whether there is a potential concern with operation of an application. Based on the determination that there is a potential concern with operation of an application (e.g., based on an analysis of a reduction of a particular type of interactions with the application), a trigger (e.g., an indication) may be generated to further investigate operation of the application. The system and method disclosed herein may then perform sentiment and NLP analysis on consumer data including application store feedback data to determine a sentiment of user comments in the application store feedback data, and to further determine a function associated with the application store feedback data. The system and method disclosed herein may then analyze user interaction data with the application to verify the sentiment and NLP analysis on the consumer date. The system and method disclosed herein may further correlate the analysis of the user interaction data with application code development data to generate a defect ticket. The defect ticket may include the particular context of the defect, and an indication of an assignment to an appropriate entity that is to resolve the particular defect. Based on the generation of the defect ticket, the system and method disclosed herein may modify the code of the application to resolve the defect.
The multi-data analysis based proactive defect detection and resolution system and the method for multi-data analysis based proactive defect detection and resolution disclosed herein provide a technical solution to technical problems related, for example, to defect detection and resolution. The system and method disclosed herein provide the technical solution of a data analyzer that is executed by at least one hardware processor (or the data analyzer may be designated as a hardware implemented data analyzer), to analyze operational data for an application to determine whether a functionality related to the application is below a predetermined threshold associated with the functionality related to the application. In response to a determination that the functionality related to the application is below the predetermined threshold associated with the functionality related to the application, the data analyzer may generate an indication to perform defect analysis related to the functionality related to the application. Further, the data analyzer may perform, in response to the generated indication, a sentiment analysis on consumer data related to the application to determine a sentiment of the consumer data related to the application, and a NLP analysis, in response to a determination that the sentiment is a negative sentiment, on the consumer data related to the application to determine a function associated with the negative sentiment. A defect detector that is executed by the at least one hardware processor (or the defect detector may be designated as a hardware implemented defect detector) may analyze, in response to the determination that the sentiment is the negative sentiment, application code and process data related to the application to determine a defect associated with the application by comparing a new user interaction pattern with the application to a previous user interaction pattern with the application, and in response to a determination that the new user interaction pattern with the application is different from the previous user interaction pattern with the application, identifying the defect associated with the application. Further, a defect resolver that is executed by the at least one hardware processor (or the defect resolver may be designated as a hardware implemented defect resolver) may modify a code of the application to correct the defect associated with the application. A defect ticket generator that is executed by the at least one hardware processor (or the defect ticket generator may be designated as a hardware implemented defect ticket generator) may determine a context of the defect associated with the application, determine an entity associated with development of the function associated with the negative sentiment, and generate a defect ticket that includes an indication of the context of the defect associated with the application, and the entity associated with development of the function associated with the negative sentiment.
A defect detector 112 that is executed by the at least one hardware processor may analyze, in response to the determination that the sentiment is the negative sentiment, application code and process data 114 related to the application 106 to determine a defect associated with the application 106 by comparing a new user interaction pattern with the application 106 to a previous user interaction pattern with the application 106, and in response to a determination that the new user interaction pattern with the application 106 is different from the previous user interaction pattern with the application 106, identifying the defect associated with the application 106.
A defect ticket generator 116 that is executed by the at least one hardware processor may determine a context of the defect associated with the application 106, determine an entity associated with development of the function associated with the negative sentiment (e.g., based on application code development data 118), and generate a defect ticket 120 that includes an indication of the context of the defect associated with the application 106, and the entity associated with development of the function associated with the negative sentiment.
A defect resolver 122 that is executed by the at least one hardware processor may modify a code of the application 106 to correct the defect associated with the application 106.
Referring to
Referring to
Based on the analysis of the operational data 104, if a number of the transactions are below a predetermined threshold for a specified amount of time, at 402, a trigger (e.g., an indication) may be generated by the data analyzer 102 for further analysis of the application-related data 108.
For the example of
Based on the generation of the trigger at 402, at 406, the data analyzer 102 may analyze the application code and process data 114 and the consumer data 110. For example, the application code and process data 114 may include machine readable instructions and process data related to the application 106, and the consumer data 110 may include application store feedback data. In this regard, the data analyzer 102 may perform a sentiment and functional analysis on the consumer data 110. The sentiment analysis on the consumer data 110 may be used, for example, to determine a sentiment (e.g., a positive or negative sentiment) of the application store feedback data (i.e., a sentiment of the text of the application store feedback data). The functional analysis may be used to determine a function associated with the application 106 that is being addressed in the application store feedback data. For example, the function associated with the application 106 may pertain to logins related to the application 106. The functions associated with the application 106 may be ascertained based on a comparison of language of the application store feedback data (e.g., based on a NLP analysis of the application store feedback data) to a catalog of known functions (e.g., login, logout, etc.) associated with the application 106. The catalog of the known functions may be described as a set of tags or keywords that are associated with known functions associated with the application 106, where the tags or keywords may be matched to the language of the application store feedback data. Thus, the functionality analysis of the application store feedback data may be used to determine the cause of the sentiment of the application store feedback data.
For the example of
At 408, the defect detector 112 may further analyze the application code and process data 114 to determine a defect associated with the application 106. For example, the defect detector 112 may further analyze user application interaction data related to the user interaction with the application 106. For example, the defect detector 112 may compare the typical user interaction pattern before the release of the application 106 to the user interaction pattern after the release of the application 106. Based on this analysis, the defect detector 112 may determine that the user interaction is irregular (e.g., broken), for example, at the “continue” screen (i.e., the user cannot perform an action that they were previously able to perform, or should be able to perform).
At 410, the defect ticket generator 116 may determine an entity (e.g., person, company, team, etc.) that worked on the particular application 106, and/or the particular defective functionality associated with the application 106. For the example of
The defect resolver 122 may modify code (i.e., machine readable instructions) of the application 106 to correct the defect associated with the application 106.
For example, the defect resolver 122 may localize the defect within a single file of a module. Based on static analysis of the file, functionality implemented in that file and a knowledge base of known machine readable instruction patterns, the defect resolver 122 may correct a defect associated with the application 106. For example, the defect resolver 122 may correct a defect associated with the application 106 if the application 106 uses a known application programming interface (API) for invoking a web service for authentication. Static analysis of the file invoking this API may show that the parameters passed to the web service are inaccurate, for example there is an error in the uniform resource locator (URL) of the web service. The defect resolver 122 may analyze and transform the machine readable instructions, and change the URL specified in the defective machine readable instructions.
After making a change, the defect ticket generator 116 may generate a ticket and assign the task of testing the modified machine readable instructions to appropriate development and testing team members. Modification of machine readable instructions by the defect resolver 122 may reduce the time to localize, determine the defect fix, and apply the fix.
The defect resolver 122 may also generate information that provides context to developers and testers to facilitate rapid defect resolution, and/or utilize the context for defect resolution. For example, the technical context may be used to localize areas within the application 106 that have potentially led to the defect. For example, a defect reported by users through application store feedback in login functionality may tend to be localized in application modules that participate in the login process.
In order to localize defects, with respect to the catalog of functions that may be used to identify if users are reporting functional defects in their feedback, this catalog of functions may also be used to create a traceability between functions and application modules where that functionality is implemented. During application design time, this extended catalog may be generated and maintained. A form of the catalog schema that may be used to document this information is <function, {app module_1, [module_file, module_file . . . ]} . . . {app_module_n, [module_file, module_file . . . ]}. Using this information, the defect resolver 122 may reduce the application machine readable instruction exploration phase so as to focus on a much smaller subset of the application machine readable instructions for resolving a defect. Similarly, a traceability between functions, modules and test scripts may provide for the selection of a subset of test scripts for regression testing and other testing efforts.
In cases where the extended catalog maintenance is not performed, the defect resolver 122 may execute a query on the application machine readable instructions to identify modules and module files where the defective functionality may be implemented. For example, if users report a defective login functionality, the defect resolver 122 may execute a query with a single word ‘login’ over the application machine readable instructions. The defect resolver 122 may expand the query by including synonyms of the functional keyword. Synonyms may include those from a language-specific thesaurus or a domain taxonomy.
In addition to localization to modules and module files, the defect resolver 122 may also use available architectural information such as control flow graphs or data flow graphs to provide additional context.
The defect resolver 122 may be configured to operate on multiple information sources, such as, for example, an extended functional catalog, a domain taxonomy, flow graphs, etc. In this manner, the capabilities of the defect resolver 122 may be extended or configured for each application.
Referring to
Referring to
Referring to
Referring to
Referring to
The release metric data may also be used to assign different criticality levels to a particular defect determination and resolution by the defect detector 112 and defect resolver 122 as described herein. For example, if a limited number of users (e.g., 100 users) have downloaded and/or updated to a new version of the application 106, then the criticality of the particular defect determination may be lower than a scenario where a significant number of users (e.g., 10000 users) have downloaded and/or updated to a new version of the application 106.
Referring to
Referring to
With respect to the changeInAverageDownloadsPostRelease parameter, if the average downloads of a new release decrease, this parameter may operate as an indicator of poor user response and satisfaction that leads to reduced downloads.
With respect to the changeInAverageUpdatesPostRelease parameter, while the changeInAverageDownloadsPostRelease parameter captures new download activity, the changeInAverageUpdatesPostRelease parameter may capture how many users are updating their older application versions. If the general sentiment of users is poor, then users may prefer to use older application versions and wait for a more stable application release in the future.
With respect to the inactiveUsersPostRelease-inactiveUsersPreRelease parameter, this parameter may indicate how many users have stopped using the application since the current release update.
With respect to the avgTransactionsPreRelease-avgTransactionsPostRelease parameter, this parameter may capture a number of transactions (e.g., checking of account balance in a banking application) that are executed successfully.
With respect to the avgRevenuePerWeekPreRelease-avgRevenuePerWeekPostRelease parameter, this parameter may operate as a metric, for example, for revenue generation through an application.
With respect to the changeInRating parameter, this parameter may capture user sentiments through application store ratings that customers provide.
For each of the above parameters, specific thresholds may be set. Further, the priorities for each parameter may also be set. Based on the thresholds and priorities, once a metric crosses (e.g., exceeds) the threshold, the dangerScore may be updated as per the priority. For example, for a normal priority, the danger score may be incremented by one, but for a high priority, the danger score may be incremented by two. The granularity of determining these metrics may be set to a predetermined time period, such as, for example, a day, a week, a fortnight, etc., so that transient user behavior may be averaged, and the metrics may represent a consistent interpretation of quality of the application release, and acceptance by users.
The determination of the metrics may be based on release data, release transition data, user sentiment data, and other metrics/data.
The indications related to the danger score may be user configurable.
As disclosed herein with reference to
Referring to
Referring to
Referring to
Referring to
Referring to
Referring to
Referring to
Referring to
Referring to
With respect to
Referring to
Referring to
The elements of the system 100 described herein may be machine readable instructions stored on a non-transitory computer readable medium. In addition, or alternatively, the elements of the system 100 may be hardware or a combination of machine readable instructions and hardware.
Referring to
At block 2504, in response to a determination that the functionality related to the application 106 is below the predetermined threshold associated with the functionality related to the application 106, the method 2500 may include generating (e.g., by the data analyzer 102) an indication to perform defect analysis related to the functionality related to the application 106.
At block 2506, the method 2500 may include performing (e.g., by the data analyzer 102), in response to the generated indication, a sentiment analysis on consumer data 110 related to the application 106 to determine a sentiment of the consumer data 110 related to the application 106, and a NLP analysis, in response to a determination that the sentiment is a negative sentiment, on the consumer data 110 related to the application 106 to determine a function associated with the negative sentiment.
At block 2508, the method 2500 may include analyzing, by the defect detector 112 that is executed by the at least one hardware processor, in response to the determination that the sentiment is the negative sentiment, application code and process data related to the application 106 to determine a defect associated with the application 106 by comparing a new user interaction pattern with the application 106 to a previous user interaction pattern with the application 106, and in response to a determination that the new user interaction pattern with the application 106 is different from the previous user interaction pattern with the application 106, identifying the defect associated with the application 106.
At block 2510, the method 2500 may include modifying, by the defect resolver 122 that is executed by the at least one hardware processor, a code of the application 106 to correct the defect associated with the application 106.
According to examples, the method 2500 may include determining, by a defect ticket generator 116 that is executed by the at least one hardware processor, a context of the defect associated with the application 106, and an entity associated with development of the function associated with the negative sentiment, and generating, by the defect ticket generator 116, a defect ticket 120 that includes an indication of the context of the defect associated with the application 106, and the entity associated with development of the function associated with the negative sentiment.
According to examples, the method 2500 may include analyzing (e.g., by the data analyzer 102) operational data 104 for the application 106 to determine whether the functionality related to the application 106 is below the predetermined threshold associated with the functionality related to the application 106 for a predetermined time period.
According to examples, the method 2500 may include analyzing operational data 104 for the application 106 to determine whether the functionality related to the application 106 is below the predetermined threshold associated with the functionality related to the application 106 by correlating the determination that the functionality related to the application 106 is below the predetermined threshold associated with the functionality related to the application 106 to a release history of the application 106, and determining, based on the correlation, whether the determination that the functionality related to the application 106 is below the predetermined threshold associated with the functionality related to the application 106 is based on a new release version of the application 106 or a previous release version of the application 106.
According to examples, for the method 2500, the consumer data 110 related to the application 106 may include application 106 store feedback data that includes feedback from users of the application 106.
According to examples, the method 2500 may include performing (e.g., by the data analyzer 102), in response to the generated indication, the NLP analysis, in response to the determination that the sentiment is the negative sentiment, on the consumer data 110 related to the application 106 to determine the function associated with the negative sentiment by comparing a text of the consumer data 110 related to the application 106 to a catalog of known functions related to the application 106, and based on a match of the text of the consumer data 110 related to the application 106 to a catalog function from the catalog of known functions related to the application 106, identifying the catalog function as the function associated with the negative sentiment.
According to examples, the method 2500 may include performing (e.g., by the data analyzer 102), in response to the generated indication, the NLP analysis, in response to the determination that the sentiment is the negative sentiment, on the consumer data 110 related to the application 106 to determine the function associated with the negative sentiment by comparing a text of the consumer data 110 related to the application 106 to a catalog of known functions related to the application 106, and based on a match of the text of the consumer data 110 related to the application 106 to a plurality of catalog functions from the catalog of known functions related to the application 106, identifying a catalog function from the plurality of catalog functions that includes a highest number of matches as the function associated with the negative sentiment.
According to examples, the method 2500 may include analyzing (e.g., by the defect detector 112), in response to the determination that the sentiment is the negative sentiment, application code and process data related to the application 106 to determine the defect associated with the application 106 by comparing the new user interaction pattern with the application 106 to the previous user interaction pattern with the application 106 by generating a plurality of user-level graphs related to interactions of a plurality of users with the application 106, merging the plurality of user-level graphs into an application-level graph for the application 106, comparing the application-level graph for the application 106 to a previous application-level graph for the application 106 to determine a change between the application-level graphs for the application 106, and determining the defect associated with the application 106 based on the determined change between the application-level graphs for the application 106.
According to examples, the method 2500 may include modifying (e.g., by the defect resolver 122) the code of the application 106 to correct the defect associated with the application 106 by implementing an alternate functionality related to the application 106.
According to examples, the method 2500 may include modifying (e.g., by the defect resolver 122) the code of the application 106 to correct the defect associated with the application 106 by modifying code associated with the functionality related to the application 106.
According to examples, the method 2500 may include determining (e.g., by the defect resolver 122) a criticality level of the defect based on an analysis of a number of users who have downloaded a new version of the application 106 versus an overall number of users who have downloaded the application 106, in response to a determination that the criticality level of the defect is less than a predetermined criticality level threshold, assign a first time period for modifying the code of the application 106 to correct the defect associated with the application 106, and in response to a determination that the criticality level of the defect is greater than the predetermined criticality level threshold, assign a second time period for modifying the code of the application 106 to correct the defect associated with the application 106, where the first time period is greater than the second time period.
According to examples, the method 2500 may include determining (e.g., by the defect resolver 122) a criticality level of the defect based on an analysis of a number of users who have updated to a new version of the application 106 versus an overall number of users who have downloaded the application 106, in response to a determination that the criticality level of the defect is less than a predetermined criticality level threshold, assigning a first time period for modifying the code of the application 106 to correct the defect associated with the application 106, and in response to a determination that the criticality level of the defect is greater than the predetermined criticality level threshold, assigning a second time period for modifying the code of the application 106 to correct the defect associated with the application 106, where the first time period is greater than the second time period.
According to examples, for the method 2500, the application 106 may include a mobile application 106 executed in an enterprise environment by a known group of users.
According to examples, for the method 2500, the application 106 may include a mobile application 106 executed in a digital environment by an unknown group of users.
According to examples, the method 2500 may include analyzing (e.g., by the data analyzer 102) operational data 104 for the application 106 to determine whether the functionality related to the application 106 is below the predetermined threshold associated with the functionality related to the application 106 by determining a danger score that represents a level of potential issues associated with a particular release version of the application 106.
According to examples, for the method 2500, the danger score of less than a predetermined first danger score threshold may represent no potential issues with the particular release version of the application 106, the danger score of greater than the predetermined first danger score threshold and less than a predetermined second danger score threshold may represent possible potential issues with the particular release version of the application 106, and the danger score of greater than the predetermined second danger score threshold may represent definite issues with the particular release version of the application 106.
Referring to
At block 2604, the method 2600 may include analyzing, by a defect detector 112 that is executed by the at least one hardware processor, in response to the determination that the sentiment is the negative sentiment, application code and process data related to the application 106 to determine a defect associated with the application 106 by comparing a new user interaction pattern with the application 106 to a previous user interaction pattern with the application 106, and in response to a determination that the new user interaction pattern with the application 106 is different from the previous user interaction pattern with the application 106, identifying the defect associated with the application 106 based on the difference of new user interaction pattern with the application 106 from the previous user interaction pattern with the application 106.
At block 2606, the method 2600 may include modifying, by a defect resolver 122 that is executed by the at least one hardware processor, a code of the application 106 to correct the defect associated with the application 106.
According to examples, the method 2600 may include determining, by a defect ticket generator 116 that is executed by the at least one hardware processor, a context of the defect associated with the application 106, and an entity associated with development of the function associated with the negative sentiment. Further, the method 2600 may include generating, by the defect ticket generator 116, a defect ticket 120 that includes an indication of the context of the defect associated with the application 106, and the entity associated with development of the function associated with the negative sentiment.
According to examples, a method for multi-data analysis based proactive defect detection and resolution may include performing, by a data analyzer 102 that is executed by at least one hardware processor, a sentiment analysis on consumer data 110 related to an application 106 to determine a sentiment of the consumer data 110 related to the application 106, and a language processing analysis on the consumer data 110 related to the application 106 to determine a function associated with the sentiment. The method for multi-data analysis based proactive defect detection and resolution may further include analyzing, by a defect detector 112 that is executed by the at least one hardware processor, application code and process data related to the application 106 to determine a defect associated with the application 106 by comparing a new user interaction pattern with the application 106 to a previous user interaction pattern with the application 106 to identify the defect associated with the application 106. The method for multi-data analysis based proactive defect detection and resolution may further include determining, by a defect ticket generator 116 that is executed by the at least one hardware processor, a context of the defect associated with the application 106, and generating, by the defect ticket generator 116, a defect ticket 120 that includes an indication of the context of the defect associated with the application 106.
The computer system 2700 may include a processor 2702 that may implement or execute machine readable instructions performing some or all of the methods, functions and other processes described herein. Commands and data from the processor 2702 may be communicated over a communication bus 2704. The computer system may also include a main memory 2706, such as a random access memory (RAM), where the machine readable instructions and data for the processor 2702 may reside during runtime, and a secondary data storage 2708, which may be non-volatile and stores machine readable instructions and data. The memory and data storage are examples of computer readable mediums. The memory 2706 may include a multi-data analysis based proactive defect detector and resolver 2720 including machine readable instructions residing in the memory 2706 during runtime and executed by the processor 2702. The multi-data analysis based proactive defect detector and resolver 2720 may include the elements of the system 100 shown in
The computer system 2700 may include an I/O device 2710, such as a keyboard, a mouse, a display, etc. The computer system may include a network interface 2712 for connecting to a network. Other known electronic components may be added or substituted in the computer system.
What has been described and illustrated herein is an example along with some of its variations. The terms, descriptions and figures used herein are set forth by way of illustration only and are not meant as limitations. Many variations are possible within the spirit and scope of the subject matter, which is intended to be defined by the following claims—and their equivalents—in which all terms are meant in their broadest reasonable sense unless otherwise indicated.
Saxena, Amitabh, Kaulgud, Vikrant Shyamkant, Sharma, Vibhu Saujanya, Singi, Kapil
Patent | Priority | Assignee | Title |
10484429, | Oct 26 2016 | Amazon Technologies, Inc. | Automated sensitive information and data storage compliance verification |
11055090, | Aug 02 2017 | Accenture Global Solutions Limited | Component management platform |
11409590, | Apr 01 2020 | PAYPAL, INC.; PayPal, Inc | Proactive outreach platform for error resolution based on user intent in server-driven communication channels |
Patent | Priority | Assignee | Title |
7788536, | Dec 21 2004 | Citrix Systems, Inc | Automated detection of problems in software application deployments |
8363942, | Jul 13 1998 | Cognex Technology and Investment Corporation | Method for fast, robust, multi-dimensional pattern recognition |
8363972, | Jul 13 1998 | Cognex Corporation | Method for fast, robust, multi-dimensional pattern recognition |
8682097, | Feb 14 2006 | FotoNation Limited | Digital image enhancement with reference images |
9229800, | Jun 28 2012 | Microsoft Technology Licensing, LLC | Problem inference from support tickets |
9536200, | Aug 28 2013 | International Business Machines Corporation | Sentiment analysis of data logs |
20080162688, | |||
20110314438, | |||
20170046246, | |||
WO2011146750, |
Executed on | Assignor | Assignee | Conveyance | Frame | Reel | Doc |
Nov 04 2015 | SAXENA, AMITABH | Accenture Global Services Limited | ASSIGNMENT OF ASSIGNORS INTEREST SEE DOCUMENT FOR DETAILS | 044058 | /0450 | |
Nov 05 2015 | SINGI, KAPIL | Accenture Global Services Limited | ASSIGNMENT OF ASSIGNORS INTEREST SEE DOCUMENT FOR DETAILS | 044058 | /0450 | |
Nov 25 2015 | KAULGUD, VIKRANT SHYAMKANT | Accenture Global Services Limited | ASSIGNMENT OF ASSIGNORS INTEREST SEE DOCUMENT FOR DETAILS | 044058 | /0450 | |
Nov 25 2015 | SHARMA, VIBHU SAUJANYA | Accenture Global Services Limited | ASSIGNMENT OF ASSIGNORS INTEREST SEE DOCUMENT FOR DETAILS | 044058 | /0450 | |
Sep 06 2017 | Accenture Global Services Limited | (assignment on the face of the patent) | / |
Date | Maintenance Fee Events |
Sep 06 2017 | BIG: Entity status set to Undiscounted (note the period is included in the code). |
Dec 28 2022 | M1551: Payment of Maintenance Fee, 4th Year, Large Entity. |
Date | Maintenance Schedule |
Jul 09 2022 | 4 years fee payment window open |
Jan 09 2023 | 6 months grace period start (w surcharge) |
Jul 09 2023 | patent expiry (for year 4) |
Jul 09 2025 | 2 years to revive unintentionally abandoned end. (for year 4) |
Jul 09 2026 | 8 years fee payment window open |
Jan 09 2027 | 6 months grace period start (w surcharge) |
Jul 09 2027 | patent expiry (for year 8) |
Jul 09 2029 | 2 years to revive unintentionally abandoned end. (for year 8) |
Jul 09 2030 | 12 years fee payment window open |
Jan 09 2031 | 6 months grace period start (w surcharge) |
Jul 09 2031 | patent expiry (for year 12) |
Jul 09 2033 | 2 years to revive unintentionally abandoned end. (for year 12) |